The yeast Non-Mendelian factor [PSI] is thought to be self- propagating, amyloid-like aggregates of the Sup35 protein. Analogous to strains of the mammalian prion, multiple variants of [PSI] were isolated from the same genetic background. It was hypothesized that [PSI] variants are encoded by distinctive conformations of Sup35p amyloid. This proposal aim to provide evidence to support this hypothesis and to further elucidate structural differences between [PSI] variants. Sup35p mutations selectively affecting the propagation of a specific [PSI] variant will be isolated. The differential effect of these mutations on the growth of Sup35 amyloid seeds, prepared from cell extracts of different [PSI] variants, will be studied in vitro to establish a correlation with their differential effect inside the cell. Complementation mutations will be selected to identify interaction surfaces of [PSI] amyloid filaments. Insertion mutations will then be engineered to probe the secondary structure of the interaction surfaces. Combining with biochemical and biophysical characterization and electron microscopic analysis, the arrangement and structure of Sup35p monomer building blocks in distinctive [PSI] variants will be compared. Results from this investigation are likely to provide crucial information regarding the nature of prion strains and the structure of amyloid which are associated with many human disease conditions including Alzheimer's disease and Huntington's disease.